Multicenter Validation of a Deep Learning Detection Algorithm for Focal Cortical Dysplasia.

BACKGROUND AND OBJECTIVE: To test the hypothesis that a multicenter-validated computer deep learning algorithm detects MRI-negative focal cortical dysplasia (FCD). METHODS: We used clinically acquired 3-dimensional (3D) T1-weighted and 3D fluid-attenuated inversion recovery MRI of 148 patients (median age 23 years [range 2-55 years]; 47% female) with histologically verified FCD at 9 centers to train a deep convolutional neural network (CNN) classifier. Images were initially deemed MRI-negative in 51% of patients, in whom intracranial EEG determined the focus. For risk stratification, the CNN incorporated bayesian uncertainty estimation as a measure of confidence. To evaluate performance, detection maps were compared to expert FCD manual labels. Sensitivity was tested in an independent cohort of 23 cases with FCD (13 ± 10 years). Applying the algorithm to 42 healthy controls and 89 controls with temporal lobe epilepsy disease tested specificity. RESULTS: Overall sensitivity was 93% (137 of 148 FCD detected) using a leave-one-site-out cross-validation, with an average of 6 false positives per patient. Sensitivity in MRI-negative FCD was 85%. In 73% of patients, the FCD was among the clusters with the highest confidence; in half, it ranked the highest. Sensitivity in the independent cohort was 83% (19 of 23; average of 5 false positives per patient). Specificity was 89% in healthy and disease controls. DISCUSSION: This first multicenter-validated deep learning detection algorithm yields the highest sensitivity to date in MRI-negative FCD. By pairing predictions with risk stratification, this classifier may assist clinicians in adjusting hypotheses relative to other tests, increasing diagnostic confidence. Moreover, generalizability across age and MRI hardware makes this approach ideal for presurgical evaluation of MRI-negative epilepsy. CLASSIFICATION OF EVIDENCE: This study provides Class III evidence that deep learning on multimodal MRI accurately identifies FCD in patients with epilepsy initially diagnosed as MRI negative.

Automated Online Quantification Method for 18F-FDG Positron Emission Tomography/CT Improves Detection of the Epileptogenic Zone in Patients with Pharmacoresistant Epilepsy.

AIMS: To assess the validity of an online method to quantitatively evaluate cerebral hypometabolism in patients with pharmacoresistant focal epilepsy as a complement to the visual analysis of the 18F-FDG positron emission tomography (PET)/CT exam. METHODS: A total of 39 patients with pharmacoresistant epilepsy and probable focal cortical dysplasia [22 patients with frontal lobe epilepsy (FLE) and 17 with temporal lobe epilepsy (TLE)] underwent a presurgical evaluation including EEG, video-EEG, MRI, and 18F-FDG PET/CT. We conducted the automated quantification of their 18F-FDG PET/CT data and compared the results with those of the visual-PET analysis conducted by experienced nuclear medicine physicians. For each patient group, we calculated Cohen's Kappa coefficient for the visual and quantitative analyses, as well as each method's sensitivity, specificity, and positive and negative predictive values. RESULTS: For the TLE group, both the visual and quantitative analyses showed high agreement. Thus, although the quantitative analysis could be used as a complement, the visual analysis on its own was consistent and precise. For the FLE group, on the other hand, the visual analysis categorized almost half of the cases as normal, revealing very low agreement. For those patients, the quantitative analysis proved critical to identify the focal hypometabolism characteristic of the epileptogenic zone. Our results suggest that the quantitative analysis of 18F-FDG PET/CT data is critical for patients with extratemporal epilepsies, and especially those with subtle MRI findings. Furthermore, it can easily be used during the routine clinical evaluation of 18F-FDG PET/CT exams. SIGNIFICANCE: Our results show that quantification of 18F-FDG PET is an informative complementary method that can be added to the routine visual evaluation of patients with subtle lesions, particularly those in the frontal lobes.